The overall objective of this project is to eliminate the occurence of hospitalizations for patients who clinical reasons for admission are either completely preventable or modifiable by appropriate ambulatory medical care. The specific goal of the proposed research effort is to develop a valid, cost-effective method for detecting cases of preventable admission,which could be instituted in he near future and could lead to corrective action wihin the medical community. Five admission diagnosis categories with a high probability of containing preventable admission (e.g., diabetes mellitus) will be examined by two explicit-criteria quality assessment methods, utilizing different levels of information. A total of five hundred new admissions in the five diagnositic categories wll be reviewed using a detailed clinical case abstract; each case will be reviewed independently by three physicians. An analysis of the prevalence of preventable admissions and the comparative reliability and sensitivity of the two review methods will be carried out.